HEALTH SERVICES RESEARCH METHODS

HEALTH SERVICES RESEARCH METHODS

Selecting a Relevant Sample / Evaluating a Sample

Paper details Assignment Details —click here for more information on this paper Course: Research Topics in Health Administration Course Textbook: Health services research methods (2nd ed.) Author: Leiyu Shi ISBN: 978-1428352292 —Assignment Readings – • Online lectures reading (copied/pasted below – see “Online Reading”) • From your course textbook, Health services research methods (2nd ed.), read the following chapters: o Evaluation research o Design in health services research o Sampling in health services research• Use the following Web resource: o http://www.improb.com/airchives/paperair/volume2/v2i5/howto.htm —click here for more information on this paperAssignment case instructions — Important info for writer….From instructor: In my classes you will be required to rely heavily on Peer Reviewed Journals. There is a requirement that you must only utilize new research which is not older than 36 months. Failure to follow this requirement will negatively impact your grade. Additionally all case studies must include at the very least a minimum of two new references that pertain to the type of case being studied. These references must be from a PRJ and not older than 36 months. —- Your assignment consist of two parts – Question 1 & Question 2 “discussion questions” — Use “Question and Answer” format to answer all questions. – Length needed: 1-page per question, total: 2-pages. – No title page needed. — Cite two PRJ references for each question, total: 4 – use APA format. – Discussion Question 1: Selecting a Relevant Sample The research question or purpose of the study guides decisions about the research sampling. How might researchers ensure that they select a sample relevant to the research without introducing a bias? –Discussion Question 2: Evaluating a Sample Read the following two studies. Be sure to read the entire articles, not just the abstracts. – Article 1 — • Abraham, J., Sick, B., Anderson, J., Berg, A., Dehmer, C., & Tufano, A. (2011). Selecting a provider: What factors influence patients’ decision making? Journal of Healthcare Management, 56 (2), 99–114. – Read Article 1 at the following link: http://www.biomedsearch.com/article/Selecting-provider-what-factors-influence/271594409.html — Article 2 — • Chullen, C. L., Dunford, B. B., Angermeier, I., Boss, R. W., & Boss, A. D. (2011). Minimizing deviant behavior in healthcare organizations: The effects of supportive leadership and job design. Journal of Healthcare Management, 55(6), 381–397. – See Added Files for PDF copy of Article 2. — Compare the two studies by analyzing their samples. Use the following questions to guide you. 1. What sampling design is used? 2. Is the sample size adequate? 3. How does the sample affect the validity of the conclusions of the study?— “Online Reading” Week 4 OverviewWelcome to Week 4! In the first half of the course, you learned that scientific investigation proceeds by iterations of observation, explanation, and experimentation. This week, we dig deeper into one aspect of observation, sampling. Sampling is one of the techniques for gathering the data on which theories are built. Sampling is a vital part of research. It can make or mar a study. An error or bias in sampling could produce questionable data and lead to erroneous conclusions. Suppose you want to investigate the health effects of weight training. It may seem that the sample should be drawn from people who do weight training. However, you must include in your sample some people who do different types of exercise and some who do not exercise at all. Only this type of a sample will allow you to compare the health status of those who work out with weights and those who don’t. Not only are the groups considered for each research important, the sample size is also extremely important. Suppose you selected 20 individuals out of a population of 2,000. This is only 1% of the population and may not be a representative sample. Purpose of Research and SamplingA new healthcare manager at a hospital notices that the emergency room (ER) of the hospital is frequently used for nonemergency treatment. This is an inappropriate use of valuable resources from the hospital’s and the insurer’s point of view. For the patient, too, an ER visit, costing about $400, is much more expensive than a visit to a family physician. Clearly, the patients should be discouraged from visiting the ER for nonemergency treatment. What would be the best way to achieve this goal? Would patient education make a difference? The manager commissions a study to estimate the impact of patient education on use of ER for nonemergency treatment. The researcher designs a patient education program. It delivers information about proper use of ER and care options available to patients through talks and printed materials. If proven effective, the program will be delivered in the area the hospital serves through professional organizations, parent-teacher associations, and other community groups. But first, the effectiveness of the program must be tested. The purpose of this research is: Determine the effectiveness of the patient education program in lowering nonessential ER visits. The independent variable is “patient education” and the dependent variable is “ER visits.” The study will select a sample and divide it into two similar groups. The patient education program or the intervention will be delivered to one group and not to the other group. The group that receives the intervention is called the experimental group and the group that does not receive the intervention is called the control group. ER visits of both groups will be tracked over a period of 6 months following the intervention. If the number of visits for the group that received the intervention is significantly lower than the number of visits for the group that did not receive it, the intervention will be considered effective. This is a simplified description of the research study. For each element of the study, many details need to be decided. This week, we focus on the sample. The first question in sampling is: What is the population? What is the group of people from among whom the sample will be selected? In our example, all the people who live in the hospital’s encatchment area make up the population. Remember, we are interested not only in the patients who come to the ER. Anyone living in the encatchment area could potentially visit the hospital ER for nonemergency treatment. Another point to note is that the population for this study is not the market that the hospital has identified for its services. The market is made up of the people whom the hospital sees as potential customers. The population for this study is made up of people who see the hospital ER as a convenient source of care. Sample Sizeclick here for more information on this paperHow large should the sample be? A general standard is that in an experimental design, we should have at least 25 subjects per condition. We have two conditions: participation and nonparticipation in the patient education program. We need at least (25 x 2 =) 50 subjects. However, some people contacted may decline to participate in an information program. Typically less than half of the people contacted agree to participate. Therefore, the researcher needs to contact at least 50 subjects per condition. So the sample size is at least (50 x 2 =) 100. Another standard is that we need to have a larger sample size if the intervention is not expected to have a large effect. It would be safe to assume that the patient education program will produce at best a small change in the use of ER for nonemergency treatment. It is unlikely that an educational program, no matter how good it is, will bring about a large change in people’s habits of accessing care, especially when they are ill and in pain. A small change in the dependent variable may not show up in a small sample. A small sample may not show a significant statistical difference between the two groups in the study. For example, visits to ER for nonemergency treatment may decrease by 3 in the group that did not receive the intervention and by 8 in the group that did receive the intervention. Is the difference large enough to draw conclusions about the effectiveness of the intervention? For a larger sample size, differences in the two groups are more likely to be statistically significant. So it would be a good idea to at least double the number of study participants. Therefore, the sample size is now 200. Statistical formulas (to determine the sample size in an objective way) take into consideration the two main points discussed herein, namely, error rate and the expected difference between the groups. Another consideration is whether the event related to the dependent variable is rare. Though ERs of large hospitals are busy places, for an individual, a visit to ER may be rare. In this case, the number of visits tracked may be so small for both groups in the study that we can’t find a statistically significant difference between them. This problem does not require an increase in sample size, however. We can simply extend the period of study from 6 months to 8 months or 9 months to account for the effect of frequency of visits. The researcher may want to ask additional questions. For example, are uninsured patients less likely to change their ER utilization patterns in response to the intervention? If yes, is there a gender difference? Such questions add new variables to the study. In this case, the variables are insurance status and gender. If the study is to find statistically significant differences between these categories, the sample size will have to increase again. However, these variables must be considered when designing the studies and selecting the sample. The researcher would want the number of insured and uninsured to be relatively similar in both the experimental and control groups Representative SampleWhat type of subjects should be included in the sample? Researchers study a sample because it is not practical to study the whole population. But the aim is to reach a conclusion that applies to the population. The sample, then, should represent the population so that what is true of the sample may be assumed to be true of the population. Think for a moment of the opinion polls in newspapers or on television. Do the participants in these polls form a representative sample of the population? The sample is drawn from only the regular readers of one newspaper or regular viewers of one television program. Readers often choose newspapers that match their own beliefs. This certainly makes the sample non-representative. The results of an opinion poll cannot be generalized to the whole population. In an experiment, not only the whole sample but also the experimental group and the control group must be representative of the population. It may seem at first that the only way to draw a representative sample is to list hundreds of characteristics for the population and make sure that they appear in the sample. But things are not so difficult as we will see. The researcher may pick up subjects at random from the population. This is a random sample. When large enough, a random sample is representative of the population. Because subjects are not selected on any criterion, there is little danger that a group is completely left out if the sampling is done in a truly random, unbiased fashion.Sample CharacteristicsWhen subjects for a study are selected, researchers only need to take into account the characteristics of interest to the study. In the example of the ER, factors such as age, income level, education, insurance, and health status are relevant. Suppose that in the area the hospital serves, 5% of the population has high incomes and 10% of the population is over 65 years of age. If the sample has similar proportions of subjects with these and other relevant characteristics, it is a representative sample. Height, weight, color of eyes, state of origin, and hundreds of other such characteristics are not relevant to the study and need not be considered. Careful selection of variables is important to keep the sample size manageable. There may be 50 variables to describe patients’ characteristics such as age, ethnicity, gender, average salary, annual household income, self-reported health status, dietary habits, number of children, and on and on. The norm is to have 20–30 subjects per variable. If half of the 50 variables are included, the sample size may go up to (25 x 20 =) 500. Suppose the research is to be done through survey questionnaires. A typical projected response rate for a mail-in questionnaire is 30% if there is an existing relationship with respondents and 10% for “cold turkey” marketing surveys. So, for a sample size of 500—500 responses—you will have to send questionnaires to a number of people, X, such that 500 is 30% of X. To get 500 completed questionnaires, the researcher would need to send out at least 1,667 questionnaires! To design a practical study with a practical sample size, researchers have to clearly define their variables and decide which of them are essential and which are not.click here for more information on this paperResearch Topic and SamplingSo far, we have spoken of samples of people. The subject of research may, however, relate to any objects, for example, organizations, health systems, or health promotion programs. The research question and the study design determine the subjects of the sample and how they are selected. Take as an example the Waxman Report we introduced in Week 2. The study was commissioned by a policy maker in order to evaluate the scientific accuracy of the content of these educational programs. The sample in this study is not people but 13 of the most popular abstinence-only curricula. Biomedical researchers focus on how the body works. They consider the biological processes, structures, functions and mechanisms within an organism. Biological and clinical researchers focus their attention on the individual. They study the response of the body to various preventative, diagnostic, or therapeutic interventions. Public health researchers study groups of people (populations). They conduct epidemiological research, which considers the frequency, distribution, and causes of ill health. Such research studies could be retrospective, studying events that already happened or prospective, studying events as they happen. Research into health systems and health administration deals with the functioning of the health system, the costs and quality of the services provided, and the distribution of resources within the system. Samples in such research may be drawn from organizations, programs, or geographical units. A study may examine the profitability of nursing homes in rural areas. Another may assess the cost-effectiveness of preventive measures among the urban poor. Sampling and Research ConclusionThe incidence of obesity has reached epidemic proportions in the U.S. One of the causes of obesity is physical inactivity. Technological advances make much household work and other labor unnecessary. Physical activity would now come mainly from sport and other recreational activities. Are there sufficient opportunities for recreational activities? If there are, can we expect that people use them? Further, can we expect a reduction in the incidence of obesity? Google and briefly read about the HOP’N After-School Project (Dzewaltowski et al., 2010), which looks into the relationship between recreation and obesity. Recreation supply is the independent variable and healthcare expenditure is the dependent variable in this study. Recreation supply, by increasing physical activity, is expected to have an effect on health status and obesity. An interesting point here is that improved health may lead to greater physical activity. The causal relationship may run both ways. The sample in the Dzewaltowski et al. study consisted of 961 children at 8 sites. The study was originally powered to detect a .5 kg/m2 difference in BMI between a sample size of 4 intervention and 4 control schools with a reduction in the detectable difference adjusting for age, ethnicity, and gender using 20 students per group. The study explains that in each sample, there is a control site and an intervention site, with data revealed that compares the sites in the tables. Note the care taken in selecting the sample and the level of detail. This can be the standard for rigorous sampling. Keep it in mind when you evaluate research with the focus on sampling in this week’s assignments. Dzewaltowski, D. A., Rosenkranz, R. R., Geller, K. S., Coleman, K. J., Welk, G. J., Hastmann, T. J., & Milliken, G. A. (2010, December 13). HOP’N after-school project: An obesity prevention randomized controlled trial. International Journal of Behavioral Nutrition and Physical Activity, 7, 90-101